Research map / Image generation / Super-resolution
Seven Ways to Improve Example-Based Single Image Super Resolution
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Combine data augmentation, a large dictionary with hierarchical search, context reasoning, cascading, and enhanced prediction | A+ · Set5 ×3 | PSNR (dB): +0.87 | Helped |
| Increase training samples from 0.5 million to 50 million (ablation) | A+ · Set5 ×3 | PSNR (dB): 32.39dB to 32.55dB | Helped |
| Increase the training pool from 5000 to 50 million samples with 1024 regressors (ablation) | A+ · Set5 ×3 | PSNR (dB): 31.83dB ... 32.71dB | Helped |
| Increase dictionary size from 16 to 65,536 regressors (ablation) | A+ · Set5 ×3 | PSNR (dB): 32.17dB ... 32.92dB | Helped |
| Replace linear anchor search with hierarchical search (ablation) | A+A · Set5 ×3 | PSNR (dB): at most only 0.03dB | Hurt |
| Add iterative back projection (ablation) | A+ · Set5 ×3 | PSNR (dB): +0.04 | Helped |
| Add iterative back projection (ablation) | Yang · Set5 ×3 | PSNR (dB): +0.59 | Helped |
| Use a 3-stage cascade (ablation) | A+ · Set5 ×3 | PSNR (dB): 32.92dB ... 33.21dB | Helped |
| Add enhanced prediction by averaging transformed input predictions (ablation) | A+ · Set5 ×3 | PSNR (dB): +0.09 | Helped |
| Add enhanced prediction by averaging transformed input predictions (ablation) | ANR · Set5 ×3 | PSNR (dB): +0.05 | Helped |
| Add context-specific regressors (ablation) | A+ · Set5 ×3 | PSNR (dB): +0.12 | Helped |
| Use an internal dictionary instead of an external dictionary (ablation) | example-based SR · L20 ×3 | PSNR gain over external dictionary | Mixed |
Closest papers by meaning
- Robust Reference-Based Super-Resolution With Similarity-Aware Deformable Convolution (CVPR 2020 · similarity 0.88)
- Transcending the Limit of Local Window: Advanced Super-Resolution Transformer with Adaptive Token Dictionary (CVPR 2024 · similarity 0.87)
- Learning Coupled Dictionaries from Unpaired Data for Image Super-Resolution (CVPR 2024 · similarity 0.86)
- SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolution (ICLR 2024 · similarity 0.86)
- Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning (CVPR 2024 · similarity 0.86)
- See More Details: Efficient Image Super-Resolution by Experts Mining (ICML 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.